Austria

Recent advancements in state-of-the-art (SOTA) Large Language Model (LLM) agents, especially in multi-turn dialogue tasks, have been primarily driven by supervised fine-tuning and high-quality human feedback. However, as base LLM models continue to improve, acquiring meaningful human feedback has become increasingly challenging and costly. In certain domains, base LLM agents may eventually exceed human capabilities, making traditional feedback-driven methods impractical. In this paper, we introduce a novel self-improvement paradigm that empowers LLM agents to autonomously enhance their performance without external human feedback. Our method, Juxtaposed Outcomes for Simulation Harvesting (JOSH), is a self-alignment algorithm that leverages a sparse reward simulation environment to extract ideal behaviors and further train the LLM on its own outputs. We present ToolWOZ, a sparse reward tool-calling simulation environment derived from MultiWOZ. We demonstrate that models trained with JOSH, both small and frontier, significantly improve tool-based interactions while preserving general model capabilities across diverse benchmarks. Our code and data are publicly available on GitHub.

ACL 2025

Sparse Rewards Can Self-Train Dialogue Agents

toolwoz

josh

tool-calling

preference tuning

supervised finetuning

user simulation

task oriented dialogue

algorithm

simulation

synthetic data

beam search

self-training

benchmark

dialogue

dataset

### Welcome!
The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024) will take place in Bangkok, Thailand from August 11th to 16th, 2024. Our Virtual Poster Sessions will take place online Thursday, August 22, 2024.

You are required to register for this event. **Please register [here](https://2024.aclweb.org/registration). **

If you have already registered, please check your inbox for an email from Underline granting you access to ACL 2024 content.

Please register!

ACL 2024

The 62nd Annual Meeting of the Association for Computational Linguistics (ACL 2024) will take place in Bangkok, Thailand from August 11th to 16th, 2024. More information will be announced soon.

poster

### Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Message from the General Chair: 
*It is my great pleasure and honor to welcome you to the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025), held in beautiful Vienna, Austria, from July 27 to August 1, 2025. ACL2025continues our field’s tradition of excellence in scholarship, innovation, and inclusivity, and I am deeply grateful to the many volunteers who have worked tirelessly to bring this event to life.* 
[Read more](https://drive.google.com/file/d/1GI_hvOpjswAuYdUTromfeDiPpCcqidwg/view?usp=sharing)

To access this event page, you need to log in with the **email address you registered with**. Access credentials will be sent to your email from Underline - subject line "Welcome to ACL 2025". Please be sure to check your spam email folder if you do not see an email confirmation right away.

Please log in

To access this event page, you need to log in with the **email address you registered with**. 

Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Despite the widespread use of 'artificial intelligence' (AI) framing in Natural Language Processing (NLP) research, it is not clear what researchers mean by ''intelligence''. To that end, we present the results of a survey on the notion of ''intelligence'' among researchers and its role in the research agenda. The survey elicited complete responses from 303 researchers from a variety of fields including NLP, Machine Learning (ML), Cognitive Science, Linguistics, and Neuroscience.
We identify 3 criteria of intelligence that the community agrees on the most: generalization, adaptability, \& reasoning.
Our results suggests that the perception of the current NLP systems as ''intelligent'' is a minority position (29%).
Furthermore, only 16.2% of the respondents see developing intelligent systems as a research goal, and these respondents are more likely to consider the current systems intelligent.

Research Community Perspectives on "Intelligence" and Large Language Models

In this paper, we propose the first multilingual study on definition modeling. We use monolingual dictionary data for four new languages (Spanish, French, Portuguese, and German) and perform an in-depth empirical study to test the performance of pre-trained multilingual language models on definition modeling of monosemic words when finetuned on this data. Furthermore, we use a zero-shot approach to test the multilingual capabilities of two popular chat-based Large Language Models (LLMs) in the task. Results show that multilingual language models can perform on-pair with English but cannot leverage potential cross-lingual synergies, with LLMs generally offering better performance overall. A comprehensive human evaluation of the LLM-generated definition highlights the zero and few-shot capabilities of these models in this new task, also showing their shortcomings. Finally, we show that performance on our task via BERTScore strongly correlates to the performance on multilingual LLM benchmarks, suggesting that our task offers a viable compute-constrained, stable and natural alternative to these.

Multilingual Definition Modeling

Prompt-based text embedding models, which generate task-specific embeddings upon receiving tailored prompts, have recently demonstrated remarkable performance. However, their resulting embeddings often have thousands of dimensions, leading to high storage costs and increased computational costs of embedding-based operations. In this paper, we investigate how post-hoc dimensionality reduction applied to the embeddings affects the performance of various tasks that leverage these embeddings, specifically classification, clustering, retrieval, and semantic textual similarity (STS) tasks. Our experiments show that even a naive dimensionality reduction, which keeps only the first 25% of the dimensions of the embeddings, results in a very slight performance degradation, indicating that these embeddings are highly redundant. Notably, for classification and clustering, even when embeddings are reduced to less than 0.5% of the original dimensionality the performance degradation is very small. To quantitatively analyze this redundancy, we perform an analysis based on the intrinsic dimensionality and isotropy of the embeddings. Our analysis reveals that embeddings for classification and clustering, which are considered to have very high dimensional redundancy, exhibit lower intrinsic dimensionality and less isotropy compared with those for retrieval and STS.

Redundancy, Isotropy, and Intrinsic Dimensionality of Prompt-based Text Embeddings

A dense retriever learns text embeddings to fetch relevant documents from a database in response to queries. However, real-world document streams constantly evolve, often diverging from the retriever's original training distribution. Indexing these documents without preemptive measures (e.g., updating or retraining) can lead to retrieval failures for future test queries. Hence, it is crucial to detect when to update dense retrievers before those test queries arrive, ensuring the retrieval system's maintenance. To address this challenge, we introduce a novel task of predicting whether a given corpus is out-of-domain (OOD) for a dense retriever before indexing. This task enables us to assess whether using the current retriever on the given corpus creates vulnerabilities for future test queries. We propose GradNormIR, a novel unsupervised method that leverages gradient norms to detect OOD documents within a given corpus. Experiments on the BEIR benchmark demonstrate that our method facilitates timely retriever updates in evolving corpora, providing valuable guidance for building an efficient and robust retriever system.

GradNormIR: When Should We Update the Dense Retriever in Evolving Corpora?

One-to-one tutoring is one of the most efficient methods of teaching. With the growing popularity of Large Language Models (LLMs), there have been efforts to create LLM-based conversational tutors which can expand the benefits of one-to-one tutoring to everyone. However, current LLMs are trained primarily to be helpful assistants and lack crucial pedagogical skills. For example, they often quickly reveal the solution to the student and fail to plan for a richer multi-turn pedagogical interaction.

To use LLMs in pedagogical settings, they need to be steered to use effective teaching strategies: a problem we introduce as Pedagogical Steering. We develop StratL, an algorithm to optimize LLM prompts and steer it to follow a predefined multi-turn tutoring plan represented as a transition graph.

As a case study, we create a prototype tutor for high school math following Productive Failure (PF), an advanced and effective learning design. To validate our approach in a real-world setting, we run a field study with 17 high school students in Singapore and show that StratL succeeds in steering the LLM to follow the PF tutoring strategy. Finally, we highlight challenges in Pedagogical Steering of LLMs and offer opportunities for further improvements by publishing a dataset of PF problems and our code.

Towards the Pedagogical Steering of Large Language Models for Tutoring: A Case Study with Modeling Productive Failure

In this study, we investigate whether non-English-centric large language models, `think' in their specialized language. Specifically, we analyze how intermediate layer representations, when projected into the vocabulary space, favor certain languages during generation—termed as latent languages. We categorize non-English-centric models into two groups: CPMs, which are English-centric models with continued pre-training on its specialized language, and BLMs, which are pre-trained on a balanced mix of multiple languages from scratch. Our findings reveal that while English-centric models rely exclusively on English as their latent language, non-English-centric models activate multiple latent languages, dynamically selecting the most similar one based on both the source and target languages. This also influences responses to culture difference questions, reducing English-centric biases in non-English models. This study deepens our understanding of language representation in non-English-centric LLMs, shedding light on the intricate dynamics of multilingual processing at the representational level.

What Language Do Non-English-Centric Large Language Models Think in?

Generative methods significantly advance event argument extraction by probabilistically generating event argument sequences in a structured format. However, existing approaches primarily rely on a single prompt to generate event arguments in a fixed, predetermined order. Such a rigid approach overlooks the complex structural and dynamic interdependencies among event arguments. In this work, we present GEMS, a multi-prompt learning framework that Generates Event arguments via Multi-perspective prompts and ontology Steering. Specifically, GEMS utilizes multiple unfilled prompts for each sentence, predicting event arguments in varying sequences to explicitly capture the interrelationships between arguments. These predictions are subsequently aggregated using a voting mechanism. Furthermore, an ontology-driven steering mechanism is proposed to ensure that the generated arguments are contextually appropriate and consistent with event-specific knowledge. Extensive experiments on two benchmark datasets demonstrate that GEMS achieves state-of-the-art performance, particularly in low-resource settings. The source code is available at: https://github.com/AONE-NLP/EAE-GEMS

GEMS: Generation-Based Event Argument Extraction via Multi-perspective Prompts and Ontology Steering

Theory of Mind (ToM) capabilities in LLMs have recently become a central object of investigation, sparking debates and discussions. In this position paper, we explore many lines of work in different communities in AI and cognitive science. Inspired by cognitive work, we view ToM tasks as a two-step process: (I) first, determining whether and how to invoke ToM, which includes setting the appropriate Depth of Mentalizing (DoM); and (II) second, applying correct inference given the appropriate DoM. We identify that many works about ToM in LLMs, such as benchmarks and add-on modules, tend to unjustly overlook the first step and focus exclusively on the second one, which can be framed as static logic problems. We support our empirical evidence about the difficulty of the different steps in existing benchmarks. We conclude with suggestions for improved evaluation of ToM capabilities, inspired by dynamic environments used in cognitive tasks in biological agents.

Mind Your Theory: Theory of Mind Goes Deeper Than Reasoning

Most function-calling benchmarks focus on single-turn interactions, overlooking the complexity of real-world scenarios. To quantify how existing benchmarks address practical applications, we introduce DICE-SCORE, a metric that evaluates the dispersion of tool-related information throughout the dialogue. Analyzing existing benchmarks through DICE-SCORE reveals notably low scores, highlighting the need for more realistic multi-round, multi-party scenarios. To address this gap, we present DICE-BENCH, a framework that constructs benchmark datasets by synthesizing conversations through a tool graph that maintains dependencies across rounds and a multi-agent system with distinct personas to enhance dialogue naturalness. The final dataset comprises 1,607 high-DICE-BENCH instances. Our experiments on 23 LLMs with DICE-BENCH show that significant advances are still required before such models can be deployed effectively in real-world settings.

DICE-BENCH: Evaluating the Tool-Use Capabilities of Large Language Models in Multi-Round, Multi-Party Dialogues

English-centric large language models (LLMs) often show strong multilingual capabilities. However, the multilingual performance of these models remains unclear and is not thoroughly evaluated for many languages. Most benchmarks for multilinguality focus on classic NLP tasks or cover a minimal number of languages. We introduce MEXA, a method for assessing the multilingual capabilities of pre-trained English-centric LLMs using parallel sentences, which are available for more languages than existing downstream tasks. MEXA leverages the fact that English-centric LLMs use English as a kind of pivot language in their intermediate layers. It computes the alignment between English and non-English languages using parallel sentences and uses it to estimate model performance in other languages. We conduct studies using various parallel datasets (FLORES-200 and Bible), models (Llama family, Gemma family, Mistral, and OLMo), and established downstream tasks (Belebele, m-MMLU, and m-ARC). We explore different methods to compute embeddings in decoder-only models. Our results show that MEXA, in its default settings, achieves a statistically significant average Pearson correlation of 0.90 with three established downstream tasks across nine models and two parallel datasets. This suggests that MEXA is a reliable method for estimating the multilingual capabilities of English-centric LLMs, providing a clearer understanding of their multilingual potential and the inner workings of LLMs. Leaderboard [anonymized.url] Code [anonymized.url]

Downloads

Next from ACL 2025

Research Community Perspectives on "Intelligence" and Large Language Models

Similar lecture

Bootstrapping LLM-based Task-Oriented Dialogue Agents via Self-Talk